SaaS Metric

Cohort Analysis

Definition

Cohort analysis groups customers by when they started (their cohort) and tracks a metric — usually retention or revenue — for each group over time. It reveals whether retention is improving for newer cohorts and whether retention curves flatten, which indicates a stable core of customers who stay long-term.

Formula

Period-N retention = (customers from a cohort still active in period N ÷ original cohort size) × 100

Benchmark

A healthy retention curve flattens rather than decaying to zero — the flattening level is your long-term retained base. Newer cohorts retaining better than older ones is the goal.

Why cohorts beat blended averages

A single blended churn number hides whether the product is getting better or worse at keeping customers. Cohort analysis separates each start group, so you can see if customers acquired this quarter retain better than last year — a signal a blended average erases.

The shape of the retention curve matters most. A curve that decays toward zero means you have no durable base; a curve that flattens means a stable core keeps using the product. The height of the flat part is effectively your long-term retention ceiling.

Frequently asked questions

What is cohort analysis?

Cohort analysis groups customers by their start period and tracks a metric — usually retention or revenue — for each group over time, so you can compare how different groups behave rather than relying on a single blended average.

What does a flattening retention curve mean?

When a cohort retention curve flattens instead of decaying to zero, it means a stable core of customers keeps using the product long-term. The level at which it flattens is effectively your long-term retained base.

Track this automatically

Connect Stripe and RetentionLens computes Cohort Analysis for you — with cohorts, trends and churn-risk scoring. Start on the free tier.

Benchmarks are general SaaS ranges and vary by segment, stage and business model. Last reviewed 2026-05-30.